TFLITE ENABLED WHEAT LEAF DISASE PREDICTION SYSTEM

TFLITE ENABLED WHEAT LEAF DISASE PREDICTION SYSTEM

M Alok 1, Dr. Geetha M 2

[1] Student, Department of MCA, BIET, Davangere.

[2] Associate Professor, Department of MCA, BIET, Davangere.

ABSTRACT

The TFLite-enabled Wheat Leaf Disease Prediction System represents an innovative solution leveraging TensorFlow Lite (TFLite) for efficient inference on mobile and edge devices. This system addresses the critical need for early detection and management of wheat leaf diseases, which can significantly impact crop yield and quality. By integrating deep learning models trained on wheat leaf image datasets with TFLite, the system enables real-time disease prediction directly on smartphones or edge devices without relying on cloud connectivity. This paper presents the architecture, implementation, and evaluation of the TFLite-enabled Wheat Leaf Disease Prediction System, highlighting its effectiveness in accurately identifying and classifying wheat leaf diseases. Experimental results demonstrate the system's high accuracy, low latency, and energy efficiency, making it a practical tool for farmers and agronomists to monitor and mitigate wheat leaf diseases in the field.

Keywords: AI, deep learning, Convolutional Neural Networks (CNNs), image classification, tensor flow lite.